Introduction:
For my first lab report, I chose to use ImagePlot even though in the seminar, I expressed how difficult it was to understand the technicalities of the program. However, since my final assignment will be on Duke and UNC basketball, I chose to use ImagePlot for my first lab report because I have saved many Duke basketball images and I wanted to see the comparison between the amount of Duke and UNC players in an image for a Duke vs. UNC game. As well, since it was difficult for me to learn how to use the program, I though I would take on the challenge of learning how to use it.
Process:
To help myself out with ImagePlot, I looked at the samples given when the program was originally installed. The data is in a .txt format, which can be created in Excel. For my lab, I created a data file with the following columns: Filename, Year, Royal, Baby, Month, and Title. These column choices are generic but since I wanted to compare Duke and UNC players in an image, “Royal” represents Duke and “Baby” represents UNC. Here’s a screenshot of my data file on Excel:
Figure 1: Screenshot of data file on Excel which was then saved as a .txt file.
I only chose 10 images because that is all I had for one game specifically focused on Duke and UNC.
Next, I went through the same steps that were given in the sample. But in the sample, the columns for the X and Y axis that were chosen had “Brightness” and “Saturation”. Since my data file does not have those column headings, I would not see those. Instead, I have chosen “Royal” for the X axis and “Baby” for the Y axis, as those were listed in my data file.
Figure 2: Choosing the X and Y axis of the visualization based on the column headings.
After clicking “OK”, the visualization rendered and a log file appeared. Below, you can see the visualization:
Figure 3: Visualization
The log file (download: Log.txt) records each detail about the visualization, such as the directory for the data and image files, what columns were chosen, and the colour mode.
Discussion:
Looking at the visualization screenshot, I didn’t think the visualization had completed rendering, at first glance. The visualization only shows five images when I had ten images placed in the image folder. I waited a bit longer but no more images appeared. I do not know why this problem occurred but I assume that it has something to do with the image files and data files not reading with one another.
Since I did use so little images, it is hard to see any relationship, and only five of the images showed up which doesn’t help either. Next time, I would use more than 10 images. I think at least 50 or 100 images will provide enough information.
One positive aspect of ImagePlot is that if I did have a great amount of images, you can zoom in at an exact spot to look more closely at a number of images. This can help to provide more of an analysis, especially if there is a huge chunk of data within the data file.
Conclusion:
In conclusion, I think this visualization would have been more successful if I had included more images and if the visualization didn’t show half of the images. It would be nice if the program did relay some information about what went wrong there. But, even using this program with my own data, I do not think I will be using this program for my final assignment since it cannot convey the statistical data I will be working with.
Lab 2 – Many Eyes
Introduction:
For my second lab, I used Many Eyes, created by IBM. I chose to use Many Eyes because one of the visualization types they had fitted what I envisioned for my final project. I looked at a visualization previously done on the site that was related to my topic and what I wanted to do for the final project (Figure 1).
The visualization I chose to use on Many Eyes is ‘Bubble Chart’. This visualization helps to show individual elements as well as elements a whole, which is what I wanted since I am comparing two teams to demonstrate the better team.
Figure 1: Visualization found on Many Eyes that was related to what I wanted to do for my final project.
Process:
For this lab, the data I chose is the player statistics from the 2011 ACC Men’s Basketball Championship game for Duke and UNC. But I also wanted to compare between the teams in general.
My data set is simple as I am only focusing on the points each player scored since that is easier to determine which team won. I had three headings: player, team and points. The names of each player, along with the corresponding team and points were inputted. The players inputted in the data file are the ones who played in the game, since all the players on the team do not play every game. Below is a screenshot of my Excel file with the data:
After I had inputted the data in Excel, I had to go on the Many Eyes website and input the data there. The site asks you to paste your data (Figure 3), double check that the site understands the type of values inputted in each section (Figure 4) and describe the data.
Figure 3.
Figure 4.
Once I completed all that information, I could then create the visualization. This is the result:
Discussion:
My first initial reaction when looking at the visualization was that I didn’t like the colours that displayed for the circles. They were too close in hue and it seemed as if they were all related to one another. One major element I disliked was that for UNC players, the hue was darker which made that team stand out as the “better team”, when in fact Duke is based on the total points of all the players (Figure 6). This data can be shown when the “label” and “colour” options at the bottom are placed as “player” and “team” in the respective areas.
I also didn’t like how you couldn’t change the colours. This would have improved my visualization by displaying the team that is truly better, in terms of overall points. I did like how you could highlight two different players (or more) at once to see a better comparison between players (Figure 7). If I had more data added to my data file, then this feature would be beneficial.
Figure 7.
Conclusion:
Overall, I didn’t like the visualization. It was too simple and didn’t provide much customization. It had what I wanted in general for my final project but it doesn’t provide enough features for me to add more details to the visualization.
Introduction:
For my first lab report, I chose to use ImagePlot even though in the seminar, I expressed how difficult it was to understand the technicalities of the program. However, since my final assignment will be on Duke and UNC basketball, I chose to use ImagePlot for my first lab report because I have saved many Duke basketball images and I wanted to see the comparison between the amount of Duke and UNC players in an image for a Duke vs. UNC game. As well, since it was difficult for me to learn how to use the program, I though I would take on the challenge of learning how to use it.
Process:
To help myself out with ImagePlot, I looked at the samples given when the program was originally installed. The data is in a .txt format, which can be created in Excel. For my lab, I created a data file with the following columns: Filename, Year, Royal, Baby, Month, and Title. These column choices are generic but since I wanted to compare Duke and UNC players in an image, “Royal” represents Duke and “Baby” represents UNC. Here’s a screenshot of my data file on Excel:
I only chose 10 images because that is all I had for one game specifically focused on Duke and UNC.
Next, I went through the same steps that were given in the sample. But in the sample, the columns for the X and Y axis that were chosen had “Brightness” and “Saturation”. Since my data file does not have those column headings, I would not see those. Instead, I have chosen “Royal” for the X axis and “Baby” for the Y axis, as those were listed in my data file.
After clicking “OK”, the visualization rendered and a log file appeared. Below, you can see the visualization:
The log file (download: Log.txt) records each detail about the visualization, such as the directory for the data and image files, what columns were chosen, and the colour mode.
Discussion:
Looking at the visualization screenshot, I didn’t think the visualization had completed rendering, at first glance. The visualization only shows five images when I had ten images placed in the image folder. I waited a bit longer but no more images appeared. I do not know why this problem occurred but I assume that it has something to do with the image files and data files not reading with one another.
Since I did use so little images, it is hard to see any relationship, and only five of the images showed up which doesn’t help either. Next time, I would use more than 10 images. I think at least 50 or 100 images will provide enough information.
One positive aspect of ImagePlot is that if I did have a great amount of images, you can zoom in at an exact spot to look more closely at a number of images. This can help to provide more of an analysis, especially if there is a huge chunk of data within the data file.
Conclusion:
In conclusion, I think this visualization would have been more successful if I had included more images and if the visualization didn’t show half of the images. It would be nice if the program did relay some information about what went wrong there. But, even using this program with my own data, I do not think I will be using this program for my final assignment since it cannot convey the statistical data I will be working with.
Lab 2 – Many Eyes
Introduction:
For my second lab, I used Many Eyes, created by IBM. I chose to use Many Eyes because one of the visualization types they had fitted what I envisioned for my final project. I looked at a visualization previously done on the site that was related to my topic and what I wanted to do for the final project (Figure 1).
The visualization I chose to use on Many Eyes is ‘Bubble Chart’. This visualization helps to show individual elements as well as elements a whole, which is what I wanted since I am comparing two teams to demonstrate the better team.
Process:
For this lab, the data I chose is the player statistics from the 2011 ACC Men’s Basketball Championship game for Duke and UNC. But I also wanted to compare between the teams in general.
My data set is simple as I am only focusing on the points each player scored since that is easier to determine which team won. I had three headings: player, team and points. The names of each player, along with the corresponding team and points were inputted. The players inputted in the data file are the ones who played in the game, since all the players on the team do not play every game. Below is a screenshot of my Excel file with the data:
After I had inputted the data in Excel, I had to go on the Many Eyes website and input the data there. The site asks you to paste your data (Figure 3), double check that the site understands the type of values inputted in each section (Figure 4) and describe the data.
Once I completed all that information, I could then create the visualization. This is the result:
Discussion:
My first initial reaction when looking at the visualization was that I didn’t like the colours that displayed for the circles. They were too close in hue and it seemed as if they were all related to one another. One major element I disliked was that for UNC players, the hue was darker which made that team stand out as the “better team”, when in fact Duke is based on the total points of all the players (Figure 6). This data can be shown when the “label” and “colour” options at the bottom are placed as “player” and “team” in the respective areas.
I also didn’t like how you couldn’t change the colours. This would have improved my visualization by displaying the team that is truly better, in terms of overall points. I did like how you could highlight two different players (or more) at once to see a better comparison between players (Figure 7). If I had more data added to my data file, then this feature would be beneficial.
Conclusion:
Overall, I didn’t like the visualization. It was too simple and didn’t provide much customization. It had what I wanted in general for my final project but it doesn’t provide enough features for me to add more details to the visualization.